312 
M. R. ENSIGN 
lindleyana, Nothoscordum fragrans, Citrus species, Mangifera indica, Evony- 
mus latifolius, Coelebogyne ilicifolia, Epheara sp., Cheiranthus cheiri, Vicium 
album, and certain orchids. Coulter, Barnes, and Cowles (1910) state 
that polyembryony is much more common among gymnosperms than 
among angiosperms. 
The most important question connected with polyembryony has to do 
with the origin of those plants produced in excess of the normal number 
from a single seed. When seeds normally produce but a single embryo, 
this is usually considered to arise from the fertilized egg within the embryo 
sac, being, therefore, gametic. In Citrus species, according to Strasburger 
(1878), one plant arises in this manner, while the others arise from the 
nucellar tissue. The latter, therefore, are apogamous. In other species, 
variations from this type of polyembryony are described. For example, in 
Santalum album and in Cypripedium Caleolis there are two fertilized eggs 
which develop. But in Citrus species, no other origin than that already 
noted has ever been found. This, however, does not mean that, when 
Citrus seeds produce only one plant, that one is always of gametic origin. 
In fact, Strasburger makes it clear that there is no way of distinguishing 
the gametic from the apogamous seedlings at the time of their germination 
except by sectioning. Sometimes one and sometimes the other emerges 
first, and in other cases only the apogamous plant develops. Osawa (1912) 
confirms these observations. 
In some cases a marked difference in the size of the polyembryonic 
seedlings is noted. This question s deferred for discussion later. 
Methods and Materials 
Plants. The seeds, from which the polyembryonic seedlings were grown 
for this study, were taken from grape fruit secured at the local market. 
These seeds were planted, one in each pot, and allowed to grow in the 
greenhouse. The first lot were planted by Dr. Knudson and were two years 
old at the time this study was undertaken. All of the determinations on 
the venation character were made from leaves of this lot. A second lot 
was planted later from which other determinations on relative size of plants, 
etc., were made. 
That similar environmental conditions surrounded the polyembryonic 
seedlings which were to be compared is quite well insured, since both the 
plants from a given seed germinate at the same time and develop under 
identical conditions so far as factors which may be controlled are concerned. 
One factor which undoubtedly plays an important part in determining the 
relative size of the polyembryonic plants is not controllable. It has to do 
with the size of the cotyledons with which each plant from a single seed is 
provided. This will be considered later. 
Clearing and Staining. In order to study the venation of Citrus leaves. 
